Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

From his comfortable armchair, Liam looked out at the falling snow, a steady and silent descent that softened the edges of everything. He cradled a warm mug of cocoa, the scent of chocolate and cinnamon filling the room. A gentle smile played on his lips.

His laptop chimed with a notification. The results for his literature midterm had just been posted. He opened the file. A 'D' blinked on the screen, a jarring note in the quiet room. His face fell.

He set the mug down with a thump, the tremor in his hands betraying his surprise. The falling snow, which moments before was a source of wonder, now seemed to mock his feelings.

Emotion: peaceful

Cluster: Calm / Serenity
PC1 (Valence): 2.40 Positive
PC2 (Disposition): -2.79

Role in Research

This story is one of 1,000 stories generated for the emotion peaceful. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 peaceful stories, after denoising with neutral dialogue baselines, produces the peaceful emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the peaceful vector is projected through the unembedding matrix.

Promoted:
прият0.408
disfrut0.407
gently0.390
affectionately0.379
thoughtful0.376
Suppressed:
de-0.541
S-0.493
L-0.396
'-0.388
worse-0.367